A High Tech Company’s Approach to Managing Supply Risk
The high tech industry is highly cyclical. During times of contraction, there is over-capacity. Suppliers suffer with low prices and excess inventory and often may stop investing in manufacturing capacity. Then during times of expansion, there are frequently shortages of critical components, and parts go on allocation, causing lost revenue opportunities and customer disappointment for OEMs who can’t build as many units as they have demand for.
The high tech industry has also become highly tiered, as have most manufacturing industries. This means that a shortage several tiers back in the supply chain can have a ripple effect downstream all the way to the OEMs that need those materials or components. A classic example is what happened with tantalum capacitors in 1999-2001. Although this story received lots of press, it bears repeating as a good example of what can happen.
Tantalum Tantrums – Lessons From a Shortage
Tantalum is a rare, blue-gray, very hard, transition metal where its primary use is in electronic capacitors (right). Capacitors made from tantalum have some highly desirable characteristics for use in electronics devices, such as a high capacity, long life, high reliability, and small size. For these reasons, a broad range of electronic devices, from cell phones to computers to telecoms equipment, have historically been designed using tantalum capacitors. The use of tantalum powder to manufacture tantalum capacitors represents about 60 percent of the overall worldwide consumption of tantalum.
In the late 90s, the surge in sales of electronic devices created a strong demand for tantalum capacitors that severely strained the supply of capacitor-grade tantalum powders. Global sales of cell phone handsets more than doubled in just two years from 163 million in 1998 to 356 million in 2000. To make matters worse, there were only two major tantalum powder suppliers in the world; Cabot (USA) and H.C. Starck Group (Germany). Every major producer of tantalum electrolytic capacitors obtained their refined tantalum powder from one or both of them.
Up to the 90s, tantalum had been obtained primarily as a by-product of tin mining. Shortages were already looming before this surge in demand, as tin supplies shrank. The production of ore was concentrated in a few geographies: 40% of world production came from Australia, and 80% of known worldwide reserves are in the Democratic Republic of Congo.
As a result of the convergence of all these factors (surging demand, shrinking supplies, limited processing capacity) tantalum capacitors went on allocation in early 1999 and remained so for almost two years. Lead times, which had been about 4 weeks, stretched out past 12 weeks. Even though these capacitors represented a tiny fraction of the cost of the devices they were in, they constrained production of the devices at time when demand for goods was already far outstripping production capacity. As a result, for want of a capacitor, billions of dollars of potential sales were lost.
Many manufacturers redesigned their devices to use other types of capacitors. At the same time, additional tantalum-ore production capacity came online in mid-2001, all just in time for the down-turn in the economy and a sharp decline in the demand for tantalum capacitors!
The prologue is that now, ten years later, Corporate Social Responsibility has become much more important. In particular, many high tech firms have made it a policy to avoid conflict minerals (i.e. from mines whose revenues support armed conflict and human rights abuses). One of the primary regions where conflict minerals originate is parts of the Democratic Republic of the Congo, which also happens to have by far the largest reserves of tantalum. So, it is entirely possible that we will be facing a new shortage of non-conflict-zone tantalum ore. It remains to be seen who will be the winners and losers in securing adequate supplies this time.
In 20/20 hindsight on these shortages, it is tempting to ask “couldn’t they see it coming?” As is usually the case, there were industry watchers that did predict these shortages. However, the majority of buyers at electronics firms focus on the near-term availability of components they buy directly, not the longer term availability of raw materials used far back in their supply chains. It is just not in their job description (nor practical) to be a monitor of every component or material at every node in their very complex multi-tier supply chain.
Learning the Lessons—Some Good Practices
The lessons from this shortage were not lost, however, on one high technology firm, which we will refer to simply as 'HighTechCorp.' They recognized that if their tier 1 or tier 2 vendors were unable to secure adequate supply, HighTechCorp would suffer. So, during expansion cycles, when supply markets start to tighten, HighTechCorp gets together its strategic supplier risk group of 3-4 people to look at the major commodities and consider the risks, including:
Industry capacity risk—Largely this is focused on semiconductor fabrication capacity, because of the long investments lead times required to build new fabs and ramp up capacity. HighTechCorp analyzes the chip makers’ investment plans and whether adequate investments are being made for the capacity that is anticipated. Because the semiconductor industry is so highly cyclical with huge up and down demand swings, if they miss or delay an investment, there can be a six to twelve month period of highly constrained capacity, during which there are major allocation challenges.
Geological and weather risk—This includes looking at risk for tiers further back in the chain. The Taiwanese Earthquake in 1999 did not affect HighTechCorp’s immediate, first tier suppliers. However, Taiwan made about 90% of the ASICS in the world at the time, so tier 2 was heavily constrained, and there was a severe shortage of DRAM (memory chips).
Geo-political risk—stability of the country/region is another consideration. These can be long term or temporary.
Supplier Power risk—Intel is the best example of a very powerful supplier that is usually the sole source of the primary microprocessor in most computers. Dealing with a very powerful supplier can end up with the 'tail wagging the dog'.
In selected cases, HighTechCorp secures the second tier supply based on their forecast. Then, as their first tier suppliers actually need components, HighTechCorp sells their allocation to their own suppliers. In some cases they hedge by issuing POs for options to buy, essentially buying capacity options. When their supplier needed that capacity, they would exercise the option. That approach has saved them several times.
It is of course not practical, nor necessary, for an organization to try to understand every risk, at every tier, for every component they buy, all the way back to the raw materials extracted from the ground. However, a high level analysis can be made with a modest investment, as HighTechCorp did, to identify potential trouble spots in the supply chain. Then these can be further investigated and mitigation strategies put in place as needed.
Making Sourcing Strategic
During periods of high demand and industry-wide shortages, those firms that figure out and lock up the supply of key components beforehand, will be the ones able to meet the surge in demand. While their competitors are sidelined waiting for capacity to free up, consumers will not wait, but will switch to the brand that is available, hence, giving that brand increased market share. At that point, the sourcing professional who was responsible for those smart risk mitigation decisions deserves a big promotion and a big raise! Now that’s one way to make sourcing a position ofstrategic importance.
Prior Managing Supply Risk articles in this series: